We retrospectively studied the risk factors, subtypes, and outcome of ischemic stroke in a single tertiary care center in Thailand. The mean age of our patients was 68.8 years, and 34.3% were 75 years and older. The proportion of elderly in this study is consistent with the data of general Thai population where 38.7% were elderly (10). According to the national stroke data, 31.0% of stroke patients were over 75 years old (11). The prevalence of elderly stroke patients varied among countries and studied population. A study of France found that 28% of stroke patients were elderly (12) and the absolute number of first-ever strokes increased by 47% in the patients over75 years old (12). Another hospital-based prospective study conducted in China found that 22% of the patients were over 75 years old (13) whereas a multicenter study conducted in Mexico showed that approximately 40% of stroke patients were older than 75 years (14).
Among the elderly stroke patients, hypertension and atrial fibrillation were the 2 most prominent risk factors. Hypertension was the most prevalent risk factor and documented in 73.4% of total patients in the entire cohort. This risk factor was also noted to be prevalent in many Asian studies including studies from northern China and Southeast Asia. (15,16) Our study found that 38.7% of the elderly patients had atrial fibrillation compared to only 14.3 % in younger patients. This correlated well with the stroke subtype of which the majority of the elderly patients had cardioembolism. Atrial fibrillation is strongly associated with increased age which was similar to findings reported by other studies (10,17,18,).
The most common subtype of ischemic stroke among the elderly in this study was cardioembolism which accounted for more than 40% of the cases. Atrial fibrillation was responsible for most of them. In contrast, cardioembolic stroke was responsible for only 20% of those who were younger than 75 years old. This high prevalence of atrial fibrillation may be partly related to the high rate of EKG monitoring among our stroke patients in the stroke unit. EKG monitoring is routinely performed for at least 24 hours in all stroke cases in our institution. Moreover, in those who had clinical suspicion of cardioembolic stroke, Holter monitoring was recommended for further evaluation. Alternatively, large artery atherosclerosis was relatively more common in the younger patients. In both age groups, the proportion of small vessel occlusion was not significantly different.
Regarding the outcomes, patients aged 75 years and older had 2.5 times higher mortality rate and poorer outcome than patients aged below 75 years which is consistent with previous studies (,19). The discharged NIHSS scores, mRS, and length of hospital stay were also significantly higher among the elderly. These unfavorable outcomes may be related to a greater severity of stroke among the elderly as measured by the initial NIHSS scores. Moreover, the elderly patients tended to have a higher prevalence of risk factors as well as other comorbidities such as atrial fibrillation and previous stroke. We also observed that cardioembolism was the major cause of stroke in the elderly and this stroke subtype was associated with a poor outcome. Previous studies also demonstrated that patients with cardioembolic stroke had a higher mortality rate and worse outcome (13).
Our study highlighted the importance of stroke among the elderly as we are entering ageing society (3,4). In this study, we used the age of 75 years as the cut-off point which is in line with other recent studies that focused on the treatment of the very old population (10,12). Hypertension was the most common risk factor and cardioembolic stroke was the most common stroke subtype in the elderly and was associated with poor outcome. Thus, it is very important to detect atrial fibrillation among the elderly in order to secondary prevent stroke and manage the patient appropriately.
This study had some limitations. Some confounding factors may not be fully evaluated due to the retrospective nature of the study. Second, although EKG monitoring was performed in all of the patients, however, further evaluation by Holter monitoring were only done in suspected patients. Therefore, it is possible that the number of patients who had atrial fibrillation may have been underestimated.